We have recently determined that more than 90% of melanomas overexpress the receptor tyrosine kinase (RTK) EphA2, with the EphA2 level of overexpression associated with metastic phenotype and an adverse clinical prognosis. Cross-linking of tumor cell-expressed EphA2 molecules using either ligand Ephrin A1-lg fusion proteins or activating anti-EphA2 antibodies results in RTK phosphorylation, followed by receptor internalization, c-Cbl-dependent ubiquitination and proteasome-dependent degradation. As a consequence, agonist-treated tumor cells acquire a more benign phenotype, and based on our preliminary data, they conditionally up-regulate their expression of EphA2-derived epitopes presented in MHC class I complexes. Using EphA2-specific CTL lines and clones, wehave shown in preliminary data that treatment with these EphA2 ligand agonists results in improved recognition and killing of tumor cells by anti-EphA2 CD8+ T cells in vitro and in vivo in Hu-SCID models. We have also recently observed that pharmacologic inhibition of protein tyrosine phosphatases (PTP) or HSP90 function also serves to increase EPhA2 preteasomal processing, theoretically making EphA2 peptides accessible to the MHC class I biosynthetic pathway. We hypothesize that EphA2 ligand agonists and PTP/HSP90 inhibitors may act synergistically in promoting enhanced tumor cell recognition by CTLs. As a consequence, we hypothesize that combinational immunotherapies consisting of 1) EphA2-based vaccines designed to elicit specific CTLs and 2) the conditional activation of EphA2 degradation and proteasomal processing via locoregional administration of EphA2 ligand agonists or PTP/HSP90 inhibitors will result in improved ani-melanoma efficacy. Our Specific Aims are to" Evaluate the ability of "agonists" that promote EphA2 proteasomal processing to sensitize melanoma cells to anti-EphA2 CD8+ T cell recognition in vitro (AIM 1);Test the hypothesis that combinational immunotherapies targeting the conditional proteasomal processing of EphA2 are safe and more effective than single modality therapies in mouse models in vivo (AIM 2);and Design and perform a phase I clinical trial of a combinational therapy involving DC1/EphA2 peptide immunization and conditional augmentation of tumor presentation of EphA2-derived epitopes in HLA-A2+ patients with advanced-stage melanoma (AIM 3).